Between a Rock and a Hard Place: The Tension Between Ethical Reasoning and Safety Alignment in LLMs

Computer Science > Cryptography and Security [Submitted on 4 Sep 2025 (v1), last revised 15 Apr 2026 (this version, v4)] Between a Rock and a Hard Place: The Tension Between Ethical Reasoning and Safety Alignment in LLMs Shei Pern Chua, Zhen Leng Thai, Kai Jun Teh, Xiao Li, Qibing Ren, Xiaolin Hu Large Language Model safety alignment predominantly operates on a binary assumption that requests are either safe or unsafe. This classification proves insufficient when models encounter ethical dilemmas, where the capacity to reason through moral trade-offs creates a distinct attack surface. We formalize this vulnerability through TRIAL, a multi-turn red-teaming methodology that embeds harmful requests within ethical framings. TRIAL achieves high attack success rates across most tested models by systematically exploiting the model's ethical reasoning capabilities to frame harmful actions as morally necessary compromises. Building on these insights, we introduce ERR (Ethical Reasoning Robustness), a defense framework that distinguishes between instrumental responses that enable harmful outcomes and explanatory responses that analyze ethical frameworks without endorsing harmful acts. ERR employs a Layer-Stratified Harm-Gated LoRA architecture, achieving robust defense against reasoning-based attacks while preserving model utility. Subjects: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI) Cite as: arXiv:2509.05367 [cs.CR]   (or arXiv:2509.05367v4 [cs.CR] for this version)   https://doi.org/10.48550/arXiv.2509.05367 Focus to learn more Submission history From: Shei Pern Chua [view email] [v1] Thu, 4 Sep 2025 05:53:20 UTC (2,206 KB) [v2] Fri, 12 Sep 2025 11:05:08 UTC (2,206 KB) [v3] Sat, 10 Jan 2026 05:42:21 UTC (433 KB) [v4] Wed, 15 Apr 2026 05:09:08 UTC (508 KB) Access Paper: HTML (experimental) view license Current browse context: cs.CR < prev   |   next > new | recent | 2025-09 Change to browse by: cs cs.AI References & Citations NASA ADS Google Scholar Semantic Scholar Export BibTeX Citation Bookmark Bibliographic Tools Bibliographic and Citation Tools Bibliographic Explorer Toggle Bibliographic Explorer (What is the Explorer?) Connected Papers Toggle Connected Papers (What is Connected Papers?) Litmaps Toggle Litmaps (What is Litmaps?) scite.ai Toggle scite Smart Citations (What are Smart Citations?) Code, Data, Media Demos Related Papers About arXivLabs Which authors of this paper are endorsers? | Disable MathJax (What is MathJax?)